Hybrid Indexes for Spatial-Visual Search

Abstract

Due to the growth of geo-tagged images, recent web and mobile applications provide search capabilities for images that are similar to a given query image and simultaneously within a given geographical area. In this paper, we focus on designing index structures to expedite these spatial-visual searches. We start by baseline indexes that are straightforward extensions of the current popular spatial (R*-tree) and visual (LSH) index structures. Subsequently, we propose hybrid index structures that evaluate both spatial and visual features in tandem. The unique challenge of this type of query is that there are inaccuracies in both spatial and visual features. Therefore, different traversals of the index structures may produce different images as output, some of which more relevant to the query than the others. We compare our hybrid structures with a set of baseline indexes in both performance and result accuracy using three real world datasets from Flickr, Google Street View, and GeoUGV.

Hierarchical information quadtree: efficient spatial temporal image search for multimedia stream

“…the number of query visual word in q.ψ k the number of results should be returned ω 1 the preference parameter to balance the spatial proximity, visual relevance and temporal recency ω 2 the preference parameter to balance the spatial proximity, visual relevance and temporal recency ω 3 the preference parameter to balance the spatial proximity, visual relevance and temporal recency fs(I, q) the spatial proximity between I.loc and q.loc fv(I, q) the temporal recency between I.ψ and q.ψ ft(I, q) the visual relevance between I.t and q.t fstv(I, q) the spatial temporal visual ranking score between I and q In this section, we present problem definition and necessary preliminaries of top k spatial temporal image search. Table 1 below summarizes the mathematical notations used throughout this section.…”

“…To the best of our knowledge, we are the first to study this important issue and the previous works Focused on top-k spatial image search without temporal information. The state-of-theart approaches proposed by Alfarrarjeh et al [1] employ a hybrid index that evaluate both spatial and visual features in tandem.…”